Randomized controlled trials (RCTs) have established that carotid endarterectomy (CEA) is superior to medical therapy in preventing stroke in both asymptomatic and symptomatic patients with carotid stenosis. However, while the majority of patients who undergo carotid revascularization receive treatment for asymptomatic disease, the benefit of revascularization for patients with asymptomatic disease is quite modest. In addition, the clinical trials were initiated about 20 years ago, and in the intervening period, medical therapy has greatly improved for this vascular disease due to new pharmacological advances such as high potency statins, improved antiplatelet regimens, and better control of hypertension, hyperlipidemia and diabetes. In parallel, as the surgical complication rate for CEA has decreased, the use of carotid artery stenting (CAS) has been gaining favor in the treatment of asymptomatic carotid stenosis. While recent trials have focused exclusively on comparing CEA to CAS, the declining stroke rate associated with medically managed carotid stenosis has led many to question whether either revascularization procedure is superior to aggressive vascular risk factor control. A three-arm RCT could answer this question, but such a study would be very expensive, not provide answers for a long time, and might only enroll the subset of healthiest patients typically studied in an RCT. Since the first RCT was published that demonstrated the superiority of CEA compared to medical therapy there has been a concern that the modest benefit observed in the trials may not be reproducible in the community. To address the ongoing debate of whether among patients with asymptomatic carotid stenosis, CEA is still superior to medical therapy and whether CAS is superior to medical therapy (which has never been evaluated in an RCT), we propose to examine the comparative effectiveness of initial medical therapy compared to early carotid intervention in a retrospective cohort study of 6000 patients using merged Veteran's Health Administration and Medicare data. We will specifically compare each revascularization strategy to medical therapy and also examine the impact of a number of key variables (e.g. age, presence of atrial fibrillation) on the comparative effectiveness of intervention compared to medical therapy. We will use advanced statistical techniques to overcome observed and unobserved differences in the medical therapy and intervention cohorts. We will compare the 30-day and 5 year outcomes of 3175 patients managed medically, 1500 patients who received CEA, and 1325 patients who received CAS entering the cohort from 2005 to 2008. Our overarching hypothesis is that medical therapy has improved to a degree that only certain subgroups of patients may benefit from revascularization and that for many patients medical therapy may be the preferred treatment strategy.